NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Lunar distance by camera, 1893
From: Paul Hirose
Date: 2011 Nov 12, 21:50 -0800
From: Paul Hirose
Date: 2011 Nov 12, 21:50 -0800
The 1893 annual report of the U.S. Coast and Geodetic Survey has an appendix on a test of a photographic method for determining time and longitude with an ordinary portable camera. It's a translation of an article by C. Runge of Hanover, Prussia. He thought the method might be useful to exploring parties. Sextant lunar distance observations would be replaced by a simple photographic procedure. The images could be reduced later at headquarters by a specialist. In Runge's experiment, a single plate received multiple exposures, which included 1) several instantaneous shots of the Moon, and 2) a time exposure of stars trailing through the area of the plate formerly occupied by the Moon. The camera remained stationary for the whole sequence. Watch time was recorded for each Moon exposure. During the star trail exposure, Runge made a break in the trails every 5 or 10 minutes by obstructing the lens for a few seconds. The watch time of each break was recorded. There are three ways to determine time from the photo. 1. By right ascension. A line drawn with a needle and straightedge on the emulsion, perpendicular to the star trails, represents a meridian of LHA. The watch time when a star had that LHA may be ascertained from the positions of the breaks in its trail with respect to the line. Likewise, the corresponding time for a Moon image may be calculated from its distance from the line, the more accurately if the line is very close to the limb. From the time difference, the Moon's right ascension is obtained, then converted to UT by interpolating the almanac. 2. By declination. Each star trail is a circle of known declination, so the Moon's declination at the time of an exposure may be deduced. Naturally, this method requires that the Moon have a considerable rate in declination. During the experiment it was about 15" per minute of time. The author estimated a single measurement would be good to about 20", hence roughly a minute of time, but remember an average may be taken with the multiple Moon exposures. 3. By lunar distance. A break in a star trail is effectively a star whose right ascension is the actual RA of the star, plus a correction for the elapsed time between the Moon exposure and the star trail break. The separation angle from the Moon to the "star" may be measured and reduced to UT in the usual way. In each of these methods the measured angle is topocentric, and has to be corrected to geocentric. That requires latitude and local time. In my view it rather defeats the objective, since the exploring party must still make latitude and time sights. The author tested these ideas with a single plate taken on the night of 1893 June 17 in Hanover with a 240 mm f/14 lens. He carefully set the camera on a window sill and took 8 shots of the new Moon between 2200 and 2223 local. Then from 2250 to 0040 he allowed the stars of Leo to trail through that part of the frame. With the three methods outlined above he got, respectively, 39.1, 39.1, and 38.6 minutes (mean = 38.9) for the difference between local time and Greenwich. The correct value was 38.9. Runge wrote, "This close result seems to partake of the nature of a coincidence. Still, we believe that an error of more than .2 m is precluded if the measurements on the plate are made as carefully as the one just described." He used a measuring machine (normally employed to read spectrograms) with a precision screw to position the plate under a microscope. Advantages of the Runge method: 1. No need for compromise exposure settings. The Moon shots can be optimized for a sharp limb, and the star shot made with wide open aperture and as long an exposure as desired. 2. By choosing the time of the star exposure, the star field can be shifted east or west to bring it near the Moon in the image. This lets you use a longer lens for a more favorable image scale. It also reduces the effect of refraction, since both Moon and stars have about the same altitude when photographed. Disadvantages: 1. Reduction is even more complex than a sextant lunar. 2. Camera must not move during the whole sequence of exposures. 3. The long time exposure of the stars requires frequent attention by the observer. 4. You still need a sextant for the latitude and time sights. 5. Unusable on the water. The modern hobbyist would find 2 and 3 especially irksome. Perhaps a reasonable compromise would be to get a few Moon shots, then proceed without much delay to the star shot (or vice versa). The star shot would be just long enough to form noticeable trails. (When you reach the point where trails begin to form, a longer exposure won't show dimmer stars, just longer trails.) Of course the star shot will also capture an overexposed Moon, but with planning it'll be clear of the good Moon images. With a digital camera the measuring would be easy. Even a simple program like Windows Paint lets you magnify a JPEG image and read any point's xy coordinates in pixels. Nevertheless, the lunar enthusiast of today will probably prefer to shoot lunars with a sextant. I mention this photographic method as a curiousity. "On Photography as Applied to Obtain an Instantaneous Record of Lunar Distances for Determinations of Longitude", U.S. Coast and Geodetic Survey Annual Report (1893), Appendix No. 4. http://docs.lib.noaa.gov/rescue/cgs/data_rescue_cgs_annual_reports.html Select document "1893 pt 2". The document size is 35 MB. It's a translation from the original in German, published in 1893 in Zeitschrift für Vermessungswesen, available at archive.org: http://www.archive.org/details/zeitschriftfrve23vermgoog Specifically, the following URL should open the Runge article in the online reader (I have never seen the reader to work with Internet Explorer): http://www.archive.org/stream/zeitschriftfrve23vermgoog#page/n432/mode/2up Runge was inspired by F. Stolze, "Die photographische Orstbestimmung ohne Chronometer", which book he reviewed earlier in the some volume. Unfortunately, my high school German is too rusty to get more than an occasional glimmer of sense. http://www.archive.org/stream/zeitschriftfrve23vermgoog#page/n318/mode/2up Amazingly, the Stolze book also is online at archive.org. From the U.S. Geodetic Survey document I infer Stolze explains how to use photography to get latitude, local time, and lunar distance. Runge's experiment was limited to lunar distance. http://www.archive.org/details/diephotographis00stolgoog Speaking of Runge, I believe he was none other than Carl Runge, of Runge-Kutta fame. The latter did live in Hanover at the time and was a noted spectroscopist, which would explain the spectrogram measuring machine. --